The First Universal Quantum Network

MrSeb writes "German scientists at the Max Planck Institute of Quantum Optics have created the first 'universal quantum network' that could be feasibly scaled up to become a quantum internet. So far their quantum network only spans two labs spaced 21 meters apart, but the scientists stress that longer distances and multiple nodes are possible. The network's construction is ingenious: Each node is represented by a single rubidium atom, trapped inside a reflective optical cavity. These atoms communicate with each other by emitting a single photon over an optical fiber. Each atom is a quantum bit — a qubit — and the polarization of the photon emitted carries the quantum state of the qubit. The receiving qubit absorbs the photon and takes on the quantum state of the transmitter. Voila: A network of qubits that can send, receive, and store quantum information. In another, probably more exciting test, the emitted photons were actually used to entangle the rubidium atoms."

Comments

I have no idea

[Dyinobal]

I have no idea what any of that means! or what it's ultimate implications are technologically speaking but it sounds awesome!

Anyone care to enlighten me on the subject?

Re:I have no idea

[Anonymous Coward]

It means that one day people will learn the difference between its and it's. Ah, to dream....

Re:I have no idea

[sycodon]

Really, really fast porn downloads.

Re:I have no idea

[jd2112]

It means that traceroute will be able to tell you response times or router addresses but not both.

Re:I have no idea

[Anonymous Coward]

But you'll have no way of telling whether a video is Goatse or not-Goatse until you watch it and collapse the state vector.

Re:I have no idea

[Courageous]

But you'll have no way of telling whether a video is Goatse or not-Goatse until you watch it and collapse the state vector.

Ah, Schrodinger's Goatse. Brought to you be the intersection of quantum mechanics and 4chan. Where physics and the dark, underbelly of the internet meet, even brave men fear to tread.

Entanglement Confusion

[chiddy]

FTA: In theory, entangled qubits could be the basis of a quantum network with zero latency over any distance, which would make it rather useful for the intergalactic Galnet that will eventually succeed the internet. I'm pretty sure it's impossibly to transfer information faster than the speed of light http://en.wikipedia.org/wiki/No-communication_theorem

Re:Entanglement Confusion

[SJHillman]

But without latency, what will the losing team blame it on?

Re:Entanglement Confusion

[ByOhTek]

So...

Even faster troll first pots?

Nuts to that!

Call me when we have instant transfer of data

[BenJeremy]

I'll be impressed when they figure out how to harness entangled particles to achieve instant transfer of information over vast distances.

Imagine a world with no RF generated, yet completely connected. Better yet... imagine the entire solar system or beyond connected with such a network.

Re:Call me when we have instant transfer of data

[Chris Burke]

You're wrong. Quantum entanglement does not allow any information to be transferred faster than light.

Sitting a million miles away from your partner with your entangled particle, the only thing you know is that you and your distant partner will measure a correlated result from that particle -- a fact you already knew a million years ago when you parted company in your very-nearly-light-speed ship.

You do not know, and can not control, what the value will be. You do not know if the other person has measured their particle's state or not. Measuring the state destroys the entanglement. All you know after is that the result you got will be correlated with what they get, or got.

No information transfer is possible.

However entanglement is useful for other things. Like networks where you can detect if someone snooped on your packets.

Re:Call me when we have instant transfer of data

[tnk1]

Yes, but entanglement cannot violate causality, which is basically what would happen if you transmit information faster than the speed of light. That means that entanglement itself *could* be faster than light, but it has to have some property that mangles any information you try to piggy back on the process so that it is useless as a communication source at FTL speeds.

The problem isn't getting something faster than light, it's being able to make any use of the process to transmit information.

Re:Call me when we have instant transfer of data

[Chris Burke]

How is this any different than, say, Morse code?

In Morse Code you can control what state the wire is in without destroying it before a single bit is sent. In Morse Code you can measure the output of the wire without similarly destroying it.

Observing spin does not destroy entanglement.

No. Measuring spin causes the particle to take on a definite state, breaking the entanglement, as surely as measuring it's momentum.

Even if it didn't, though, you still couldn't communicate. You'd just know that a longer sequence of spin data you saw would be correlated with the other end. Interacting with the particle to change the spin would break the entanglement and they would not see a result correlated with the spin you gave the particle.

Re:Call me when we have instant transfer of data

[canajin56]

You do not know, and can not control, what the value will be. You do not know if the other person has measured their particle's state or not. Measuring the state destroys the entanglement. All you know after is that the result you got will be correlated with what they get, or got..

You forget that quantum shit be weird. If you think of particles as particles and their state as a 0/1 variable, then that's totally true. But particles do crazy things. One of the things they do is act like waves if nobody is looking. Entangled particles have to behave like the same sort of thing. In particular, if one of them enters a two slit setup and self interferes, the entangled pair has to also act like a wave and self interfere. This apparently occurs regardless of distance. What this means is that if Alice and Bob have a shared set of atoms. If Alice shoots an atom at a pair of slits, then Bob's atom will self interfere even if shot at an unshielded detector. Now that's not useful for sending messages, because statistically Bob can't tell if it hit where it hit because it's a particle, or because it's a wave. And the quantum equations say the same thing, that statistically the two states cannot be distinguished from random noise. However, the equations do not apply to larger systems, and we don't currently have ones that do.

Now, people assume there must be some quantum effect to prevent this from being used, because superluminal signals are mutually exclusive with causality, and most people assume causality holds. But there's no strong evidence either way at the moment.

Re:Call me when we have instant transfer of data

[tendrousbeastie]

This whole conversation is based on a misunderstanding of entanglement.

Entanglement is where two (or more, but let's keep it at two) particles are created in such a way that a conservation law must be maintained - say for example a particle of spin 0 decays and emits two electrons, then these 2 electrons must have spin -1/2 and spin +1/2 (or something like that, I can't remember how conservation of angular momentum works in quantum physics, but the principle is there).

Anyway, you have two particles, whose spin must equal that of the parent particle to converse angular momentum.

So, if you measure one, you will by necessity know the spin of the other.

Quantum theory, at least in the controversial Copenhagen Interpretation, says that the particular spin values of a particle are not an inherent property of that particle, they instead are a function of its probabilistic (Schrodinger) wave function.

When you measure the particle (i.e. observer it) it then take on a spin value. To fulfil the conversation law, at the very same time the other (entangled) particle must adopt the relevant spin value that balances out.

It must take this balancing spin value immediately, even if the particles are many light years apart, otherwise there would be a creation or destruction of energy. This will happen instantly, i.e. faster than light.

So, the cliché about entanglement transmitting info is just that it tell the other particle which value to take on for one of its properties.

But, all you can do is measure the first particle. You cannot force it to have a certain spin. It will have whatever spin it is measured to have. So you cannot communicate any info. All you can do is measure you particle and know that its entangled partner will have a corresponding value.

And you can't tell that you other particle has been measured. If you measure your particle you cannot tell if you measured it first. You would get the same result if you measured it first or not.

So, entanglement does imply faster than light, but it does not allow any information to travel that we can use.

Quantum Internet

[pablo_max]

I am no physicist, so I am actually asking seriously to those of you who are.
As it is already know, particles which are entangled at the quantum level have an instant and equal reaction on one another regardless of distance. Would it not be possible to use this "Quantum Internet" for C from say, a satellite controller a rover on Mars and one on Earth?
I have heard that it is not really workable, but is that from an engineering prospective or from a laws of physics perspective?

Re:Quantum Internet

[Chris Burke]

When you change the state of one, it changes the state of another. Why could you not just view the state as a way of transferring information?

Because you can't control the state that it collapses to when you measure it and break the entanglement. You can't tell whether or not the person on the other end has already done this. All you know is that whatever state you measure, they will see a correlated result. Which you already knew; you've learned nothing.

A useful analogy* -- it's like you and the person you want to "communicate" with put two marbles, one red and one black, into two bags. You randomly pick one, your partner takes the other. You fly apart at 0.9c for a while. Then you look in your bag. It's a red marble. You now "instantly" know that your partner has a black marble -- but you haven't actually communicated anything.

* It's just an analogy; the fact that it doesn't obey Bell's theorem is immaterial to understanding why you haven't communicated anything.

Re:Quantum Internet

[Chris Burke]

So is there anything you could actually accomplish with this "network"?

Yes! Because measuring the state of the particles breaks the entanglement, in a quantum network you would theoretically be able to tell if someone was listening in -- you'd send the correlation information along with the regular data, and if it didn't line up with what you saw in your particles at your end, then you'd know that the entanglement had been broken and someone had sniffed your packets.

The application would be key exchange. You could just send a shared key over the insecure network, and if nobody intercepted it, then you're good to go.

This is how I understand it, anyway.

The article is indeed wrong about the "zero latency" aspect. If the researchers were actually claiming to be able to send information instantaneously then they'd be a lot more obvious about it -- see the recent ICARUS FTL neutrino result where the mere hint of FTL communication resulted in international attention and a constant barrage of headlines.

Heisenberg Disagrees?

[dullertap]

Doesn't the Heisenberg Uncertainty principle kind of say that when these machines are determining spins of particles, that they cause them to change? ie entropy?

Re:iPhone qubit?

[DynamoJoe]

So "Never", then?

99.8% data loss

[melonman]

From TFA, this is apparently a huge improvement on previous attempts, but it's still not exactly dazzling. What sort of self-correcting protocol do you need to handle 499 of every 500 bits being lost?

Re:99.8% data loss

[Luyseyal]

I guess we won't discuss the state-of-the-art in neutrino communication, then...

http://www.technologyreview.com/blog/arxiv/27648/

These guys used an experiment called NuMI (NeUtrino beam at the Main Injector) to generate an intense beam of neutrinos. The beam consisted of about 25 pulses each separated by 2 seconds or so, with each pulse containing some 10^13 neutrinos.

The beam is pointed at a detector called MINERvA weighing about 170 tonnes and sitting in an underground cavern about a kilometre away. To reach MINERvA, the beam has to travel through 240 metres of solid rock.

MINERvA is one of world's most sensitive neutrino detectors and yet, out of 10^13 neutrinos in each pulse, it detects only about 0.8 of them on average.

Nevertheless, that's enough to send a message. The FermiLab team used a simple on-off protocol to represent the 0s and 1s of digital code and transmitted the word "neutrino".

The entire message took about 140 minutes to send at a data rate that these guys later worked out to be about 0.1 bits per second with an error rate of less than 1 per cent.

-l

"carries the quantum state"?

[fatphil]

You can't copy quantum state. The only way it can carries the quantum state of something is if it also destroys that something's quantum state. (But of course you can't destroy quantum state either, you've effectively just swapping quantum state.)

So information might be passed around, but it's never actually being shared.
Which isn't much of a network.

Disclaimer - I'm rusty.

quantum what???

[stanlyb]

If this is really a quantum network, why do they need a fiber to send the information!!!!!!!!!!!!!

Re:iPhone qubit?

[jamiesan]

Archos is going to come out with one that has a 3 dimensional qbit array.

It will be 300 qbits long, 50 qbits wide, and 30qbits tall

Re:iPhone qubit?

[Shemmie]

I know a few Apple fans who have an iQ 4.